The Decoder· Manuel Uth·· 21 天前AI 评分62
Epoch AI 家具组装基准：GPT-6 Astra 识别宜家组装错误准确率达 80%
OpenAI's GPT-6 Astra can now tell you exactly where you screwed up your IKEA shelf
AI 导读
Epoch AI 的家具组装基准（FAB）用三件宜家家具的组装照片，让模型对照说明书找出故意设置的错误。
来源：The Decoder · the-decoder.com